An important source of xylene in an oil refinery is catalytic reformate, which is prepared by contacting a mixture of petroleum naphtha and hydrogen with a strong hydrogenation/dehydrogenation catalyst, such as platinum, on a moderately acidic support, such as a halogen-treated alumina. Usually, a C6 to C8 fraction is separated from the reformate and extracted with a solvent selective for aromatics or aliphatics to produce a mixture of aromatic compounds that is relatively free of aliphatics. This mixture of aromatic compounds usually contains benzene, toluene and xylenes (BTX), along with ethylbenzene.
However, the quantity of xylene available from reforming is limited and so recently refineries have also focused on the production of xylene by transalkylation of C9+ aromatic hydrocarbons with benzene and/or toluene over noble metal-containing zeolite catalysts. One such process, using MCM-22 as the zeolite catalyst is disclosed in U.S. Pat. No. 5,030,787. However, during the transalkylation of C9+ aromatics with, for example, toluene to produce xylem and benzene, saturated by-products, which boil in the same temperature range as the desired aromatic products, are typically produced making separation of the desired products at high purity levels difficult. For example, a commercial benzene product may need a purity of 99.85 wt % or higher. However, initial benzene purity after distillation of a transalkylation reaction product is typically only 99.2% to 99.5% due to the presence of coboilers, such as methylcyclopentane, cyclohexane, 2,3-dimethylpentane, dimethylcyclopentane and 3-methylhexane. Therefore, an additional extraction step is usually required to further improve benzene product purity to the desired level.
One solution to the problem of the production of benzene co-boilers during the transalkylation of heavy aromatics is disclosed in U.S. Pat. No. 5,942,651 and involves the steps of contacting a feed comprising C9+ aromatic hydrocarbons and toluene under transalkylation reaction conditions with a first catalyst composition comprising a zeolite having a constraint index ranging from 0.5 to 3, such as ZSM-12, and a hydrogenation component. The effluent resulting from the first contacting step is then contacted with a second catalyst composition which comprises a zeolite having a constraint index ranging from 3 to 12, such as ZSM-5, and which may be in a separate bed or a separate reactor from the first catalyst composition to produce a transalkylation reaction product comprising benzene and xylene. A benzene product having a purity of at least 99.85% may be obtained by distilling the benzene from the transalkylation reaction product, without the need for an additional extraction step. According to the '651 patent, the second catalyst composition comprises up to 20 wt % of the total weight of the first and second catalyst compositions.
Another problem associated with heavy aromatics alkylation processes is catalyst aging since, as the catalyst cokes with increasing time on stream, higher temperatures are normally required to maintain constant conversion. When the maximum reactor temperature is reached, the catalyst needs to be replaced or regenerated. Depending on the C9+:C6 or C7 composition of the feed, the cycle length may vary from only 9 months for high C9+:C7 ratios of 85:15 to about 5 years for low C9+:C7 ratios of 20:80. Recent work has shown that the aging rate of existing transalkylation catalysts is also strongly dependent on the presence in the feed of aromatic compounds having alkyl substitutents with two or more carbon atoms, such as ethyl and propyl groups. Thus these compounds tend to undergo disproportionation to produce C10+ coke precursors.
To address the problem of C9+ feeds containing high levels of ethyl and propyl substituents, U.S. Published Application No. 2009/0112034 discloses a catalyst system adapted for transalkylation a C9+ aromatic feedstock with a C6-C7 aromatic feedstock comprising: (a) a first catalyst comprising a first molecular sieve having a Constraint Index in the range of 3-12 and 0.01 to 5 wt % of at least one source of a first metal element of Groups 6-10; and (b) a second catalyst comprising a second molecular sieve having a Constraint Index less than 3 and 0 to 5 wt % of at least one source of a second metal element of Groups 6-10, wherein the weight ratio of said first catalyst to said second catalyst is in the range of 5:95 to 75:25. The first catalyst, which is optimized for dealkylation of the ethyl and propyl groups in the feed, is located in front of said second catalyst, which is optimized for transalkylation, when they are brought into contact with said C9+ aromatic feedstock and said C6-C7 aromatic feedstock in the presence of hydrogen.
Whereas the multiple catalyst bed system of U.S. Published Application No. 2009/0112034 represents a significant improvement over single catalyst bed processes for the transalkylation of heavy aromatic feeds, it suffers from the disadvantage that the first catalyst, in addition to dealkylating ethyl and propyl substituted aromatics and saturating the resultant olefins, tends to facilitate the side reaction of aromatics saturation. This reaction is primarily a concern at the start of cycle when the reactor temperature is at its lowest and is undesirable since it reduces aromatic yield and generates saturated products. In addition, some of these saturated products, e.g. cyclohexane and methylcyclohexane, have boiling points close to benzene which makes it difficult to recover high purity benzene. Although this problem can be alleviated by sulfiding the metal catalysts on start-up to reduce their aromatic saturation activity, this is often undesirable as it necessitates adding the facilities and sulfiding agents to effect the metal sulfidation.
The present invention seeks to provide a C9+ aromatic transalkylation process that retains the advantages of the multiple bed catalyst system while reducing the problem of aromatic saturation without the need for pre-sulfidation of the catalysts.